Sign up for our newsletter and get the latest HPC news and analysis.


Solving Eight Constraints of Today’s Data Center

data center cooling

With the growth of big data, cloud and high performance computing, demands on data centers around the world are expanding every year. Unfortunately, these demands are coming up against significant opposition in the form of operating constraints, capital constraints, and sustainability goals. In this article, we look at 8 of these constraints and how direct-to-chip liquid cooling is solving them.

Poznan in Poland Goes Green with CoolIT Systems

poznan

Today CoolIT Systems announced that the Poznan Supercomputing and Networking Center has successfully deployed Rack DCLC liquid cooling. Through a collaboration with Huawei and Itprojektof, the company described the new installation as the most energy efficient HPC cluster in Poland.

PNNL Looks at Undervolting to Meet Exascale Goals

Figure depicting the general constraints at scale for building exascale architectures. The undervolting component (shown in 4) is part of ongoing work that PNNL’s HPC group and their collaborators are conducting in the effort to build highly resilient and energy-efficient HPC systems.

PNNL researchers are using supercomputers to take on two of the main challenges of exascale: energy efficiency and resiliency. Their simulations show that dynamic voltage scaling, also known as undervolting, can reduce power consumption and leverage existing mainstream resilience techniques at scale for improving system failure rates.

Hunting Supernovas with Supercomputers

Simulation of the expanding debris from a supernova explosion (shown in red) running over and shredding a nearby star (shown in blue).

Caltech researchers are using NERSC supercomputers to search for newly born supernovas. The details of their findings appear May 20 in an advance online issue of Nature.

Video: Accelerated Quantum Chemistry with CP2K

cp2k

“Learn how we achieve great GPU performance with an auto-tuned sparse matrix multiplication library, enabling quantum simulation of millions of electrons.”

Video: High Performance Computing with Python

sharcnet

“This talk will discuss various strategies to make a serial Python code faster, for example using libraries like NumPy, or tools like Cython which compile Python code. The talk will also discuss the available tools for running Python in parallel, focusing on the mpi4py module which implements MPI (Message Passing Interface) in Python.”

Interview: HPC Success Stories Session Coming to ISC 2015

Frank Baetke, HP

This year, the ISC High Performance Conference will feature a new session called Great Success Stories of HPC. To learn more, we caught up with session chair Frank Baetke from HP.

Sandia’s Mark Taylor Receives DOE Secretary’s Honor Award

Sandia National Laboratories’ Mark Taylor is the chief computational scientist for the Department of Energy’s Accelerated Climate Modeling for Energy executive council team.

Today Sandia National Laboratories announced that researcher Mark Taylor has received the U.S. Department of Energy (DOE) 2014 Secretary’s Honor Award — the department’s highest non-monetary employee recognition — for his work as chief computational scientist for DOE’s Accelerated Climate Modeling for Energy (ACME) executive council team.

New Magazine Celebrates PRACE Women in HPC

WomenofHPC2015

The Partnership for Advanced Computing in Europe in Europe has published the first edition of PRACE Women in HPC Magazine, a collection of success stories celebrating the contribution women make to HPC and computational science. Designed to put the spotlight on the scientific advances made in the past year, the magazine tells the story of women who help make PRACE a world-leading force in HPC-enabled science and the march towards Exascale computing.

GCS in Germany Awards Close to 1 Billion Core Hours

logo

Today the Gauss Centre for Supercomputing (GCS) in Germany announced the award of close to 1 billion compute core hours to scientifically outstanding national research projects.